Some a bit incorrect color and shade gradations

[Kouzeru]

I'm putting some sort of image test to display some hue and value spectrum,
I got some noticeable color banding when using this driver:

I had deployed another technique to actually do the BCM properly, thus, displaying more faithful hue and value spectrum:

(with the cost of sacrificing 1/4 of the total brightness of course, due to limitation of my driver)

Here;s the code of the test image

  char t = millis()>>4;
  for(int x=0;x<128;x++)
    for(int y=0;y<64;y++)
      dma_display->drawPixelRGB888(x,y,
      (uint8_t)((max(min(cosf(2.f*PI*((x-t)/128.f+0.f/3.f))+0.5f,1.f),0.f)*((y+1.f)/64.f))*255.f),
      (uint8_t)((max(min(cosf(2.f*PI*((x-t)/128.f+1.f/3.f))+0.5f,1.f),0.f)*((y+1.f)/64.f))*255.f),
      (uint8_t)((max(min(cosf(2.f*PI*((x-t)/128.f+2.f/3.f))+0.5f,1.f),0.f)*((y+1.f)/64.f))*255.f)
      );

[mrcodetastic]

Can you confirm if you get this issue when using version 3.0.0 of the library?

I may have introduced a bug when I fiddled with the way brightness is managed in the most recent version

[Kouzeru]

I had tested the version 3.0.0; it still yields the exact same result:

But I'd think I could try reproduce how your driver translate the colors:

which was plotted pixel by pixel in python code

u -= 2
r = int((max(min(sin((u/8+0/3)*pi*2)+0.5,1),0)*(1-v)/2)*255)
g = int((max(min(sin((u/8+1/3)*pi*2)+0.5,1),0)*(1-v)/2)*255)
b = int((max(min(sin((u/8+2/3)*pi*2)+0.5,1),0)*(1-v)/2)*255)
r0,r1,r2,r3,r4,r5=r>>7&1,r>>6&1,r>>5&1,r>>4&1,r>>3&1,r>>2&1
g0,g1,g2,g3,g4,g5=g>>7&1,g>>6&1,g>>5&1,g>>4&1,g>>3&1,g>>2&1
b0,b1,b2,b3,b4,b5=b>>7&1,b>>6&1,b>>5&1,b>>4&1,b>>3&1,b>>2&1
r,g,b=r0*4+r1*2+r2+r3+r4,g0*4+g1*2+g2+g3+g4,b0*4+b1*2+b2+b3+b4
sqrt(r/9)*256,sqrt(g/9)*256,sqrt(b/9)*256

[Kouzeru]

The pull request #374 will fix the following artifacts


into

And I already had thorough tests, and only this is was complaint of the driver.